When you think of data footprint constraints in enterprise data storage, what typically comes to mind are physical...
cabinets, racks and floor space. But there are other footprint constraints often found in storage infrastructures, including availability, capacity, power and cooling, license and maintenance fees, management tools, staffing and budgets.
To illustrate a point, let's take a look at a generic scenario of a mixed workload environment, before and after performing a data footprint optimization. When contemplating these scenarios, you should scale performance and capacity up or down to meet the specific needs of your environment.
The before example consists of two storage systems, one configured for high-performance and another configured for high-capacity secondary storage. The after example shows a single storage system configured with a mix of both high-performance and high-capacity storage.
For the performance storage system, 192 146 GB 15K rpm (20 TB raw) disks provide good performance, albeit with low capacity utilization. This translates into a low capacity-per-watt value but provides reasonable IOPS per watt and some performance hot spots. The combined energy use of both arrays is about 15,000 watts, which translates to approximately $16,000 in annual energy costs (cooling excluded) when assuming an energy cost of 12 cents per kWh.
For the capacity-centric storage system, there are 192 1 TB disks (192 TB raw) with good space utilization. However, there are some storage performance bottlenecks, including growth constraints and a low IOPS per watt.
By using a combination of techniques, net performance, capacity and feature functionality can be increased, while floor space, power, cooling and associated footprints can be reduced.
By moving heavily accessed files or data, essentially consolidating I/Os to faster, yet higher utilized solid-state drives (SSD) or 15.5K disks, overall net capacity utilization can go up without impacting service quality.
Specifically, using a mix of technologies aligned to meet specific tasks provides a balance of performance, availability, capacity and energy. It can save costs, enable growth and allow another storage system to be put into that footprint (floor space, power, cooling, operating costs).
Data storage tools to reduce your data footprint
Storage tools you should be leveraging include storage resource management (SRM) tools, compression techniques, massive array of idle disks (MAID), data deduplication, thin provisioning and SSDs and other tiered storage technologies.
In the quest to reduce your data footprint, keep in mind that what might be applicable to one environment or application may not apply to another. Here are some final tips for reducing your data footprint:
- Establish a performance and capacity baseline
- Align applicable RAID level configuration to meet requirements
- Deploy disks drives of the right size and performance to meet your specific needs
- Balance performance and capacity optimization to your particular environment
- Gain insight into how resources are used to deliver a level of service
- Leverage data footprint reduction tools for online and offline storage